Zika virus (ZV) is a mosquito borne flavivirus (FV) with unique properties that permit it to persist in patient blood and urine for months, to be sexually transmitted, and to cause encephalitis and fetal microencephaly. In patients, ZV infects endothelial cells (ECs) and crosses the EC lining of capillaries, which normally separate maternal and fetal blood supplies, resulting in transplacental transmission and infection of fetal tissues. How ZV persists and crosses brain and placental barriers remains to determined. We recently found that ZV persistently infects and replicates in primary human cerebral MECs (CMECs) and human umbilical vein ECs (HUVECs). This target provides the potential for ZV to reside in patient ECs, disseminate in bodily fluids and concomitantly bypass placental and brain capillary barrier functions. Here we investigate the mechanism by which ZV establishes persistence and spread within primary human ECs. The ability of ZV and other FVs to infect cells is acutely sensitive to the prior addition of Type I interferon (IFN?/?). In mice deficient in IFN?/? receptors, ZV infects placental and fetal brain tissues resulting in fetal demise that mimics human microencephaly. In IFN locus deficient Vero cells, ZV replicates to high titers, causing a self-limiting infection that is cytopathic, similar to the ZV effects on neurons. Cytotoxicity is consistent with acute transient infections caused by other FVs, but is contrary to findings of ZV persistence in patients. Distinct from Veros, we observed no cytopathology associated with ZV (PRV) infection of primary human CMECs or HUVECs. We found that ZV efficiently infected ECs, replicated to high titers and inhibited the induction of IFN? and IFN stimulated genes 12- 48hpi. Remarkably, we observed that ZV continued to replicate and spread in ECs, even after passaging ZV infected CMECs or HUVECs serially. This does not occur in dengue virus infected ECs where late IFN? induction restricts viral spread. These findings suggest that ZV uniquely regulates and persists within primary human ECs. In addition to IFN inhibition, our transcriptional analysis of ZV infected CMECs revealed 2 novel responses critical to EC and neuronal function. We found that ZV highly induced TRAIL/Apo2 and Rnd1 in infected CMECs. TRAIL is a secreted ligand that causes neuronal apoptosis and Rnd1 is a Rho family GTPase that inhibits axonal extensions linked to human cognitive deficits. However, in ECs TRAIL uniquely directs cell survival and proliferative Akt and ERK responses that protect ECs from apoptosis, and Rnd1 is a novel, constitutively active, inhibitor of inter-endothelial cell adherence that causes permeability. These findings suggest novel mechanisms for ZV to persist in endothelial cells and cross fetal and cerebral capillary barriers. Persistently infected ECs may serve as a source of ZV spread, persistent viremia (1-3 months) and capillary permeability that permits ZV access to fetal and neuronal tissue. Thus ZV infection of ECs may play a strategic role in ZV persistence and spread. My lab investigates EC permeability directed by RNA viruses, mechanisms of IFN regulation and therapeutic targeting of pathways to restore EC barrier functions. Here we analyze responses of primary human ECs to ZV infection. We define roles for IFN, TRAIL and Rnd1 in ZV persistence, cell survival and increased EC permeability, that in vivo may contribute to person to person and tissue specific ZV transmission. Analysis of ZV persistence in ECs may reveal therapeutic approaches to clear ZV infections, enhance EC barrier functions and reduce ZV access to fetal and neuronal tissues. !